Jump to content

Nvidia reveals key GP100 Pascal details at GTC 2016 (610mm2, 15.3B transistors, 300W TDP, 3584 cores, 1480MHz boost clock)

tareyza
By tareyza
in Tech News
 Share
Posted

 

Quote

 

SSP_79_575px.JPG

 

01:24PM EDT - 150B Transistors

01:23PM EDT - "The most ambitious project we have ever undertaken"

01:23PM EDT - Tesla P100

01:23PM EDT - writeln ('Hello, world.')

01:23PM EDT - Pascal time!

SSP_80_575px.JPG

 

01:34PM EDT - 170TF FP16 in a box. 8 P100s in a hybrid cube mesh

01:34PM EDT - A full deep learning rackmount server

01:34PM EDT - NVIDIA DGX-1

01:33PM EDT - But if it's 600mm2 for just the die, that's a huge jump in the size of dies being produced on 16nm/14nm TSMC/Samsung FinFET

01:33PM EDT - Need to get confirmation on whether 600mm2 is just the GPU die, or if they're counting other parts as well

01:32PM EDT - "We'll ship it... soon"

01:32PM EDT - P100 servers coming in Q'17

01:31PM EDT - P100 in volume production today

01:30PM EDT - NV wanted new algorithms to take advantage of the hardware

01:29PM EDT - Recapping NVLink. 5x the aggregate speed of PCIe 3.0

01:29PM EDT - "TSMC CoWoS® (Chip-On-Wafer-On-Substrate) services use Through Silicon Via (TSV) technology to integrate multiple chips into a single device. This architecture provides higher density interconnects, decreases global interconnect length, and lightens associated RC loading resulting in enhanced performance and reduced power consumption on a smaller form factor."

01:28PM EDT - Chip on Wafer on Substrate, the largest such chip ever made

01:28PM EDT - Er, 600mm^2 !!!!

01:27PM EDT - 500mm^2 !!

 

 

Note: Anandtech's original live blog was written with newest updates on top, so read the post down to up, not up to down like how one normally reads posts. 

 

Key details: 

5.3TF FP64, 10.6TF FP32, 21.2TF FP16, 14MB SM Register File, 4MB L2 Cache

 

Anandtech's live GTC blog of Nvidia's keynote address reveals several interesting details about GP100, the big Pascal die going into their new Tesla GPU. Notably, it features a very large 14MB register file, has FP64 compute as 1/2 that of FP32 compute, and has a 600mm^2 die. (The 150 billion transistors mentioned above includes transistors found in RAM). Anandtech also speculates that Pascal may be released in June, to coincide with the launch of CUDA 8 which supports Pascal. 

 

Source: Anandtech (http://www.anandtech.com/show/10225/the-nvidia-gtc-2016-live-blog)

 

Updates: NVIDIA has also posted a detailed article on GP100 on their website, with key details shown below. 

 

Quote

Like previous Tesla GPUs, GP100 is composed of an array of Graphics Processing Clusters (GPCs), Streaming Multiprocessors (SMs), and memory controllers. GP100 achieves its colossal throughput by providing six GPCs, up to 60 SMs, and eight 512-bit memory controllers (4096 bits total). The Pascal architecture’s computational prowess is more than just brute force: it increases performance not only by adding more SMs than previous GPUs, but by making each SM more efficient. Each SM has 64 CUDA cores and four texture units, for a total of 3840 CUDA cores and 240 texture units.

 

Delivering higher performance and improving energy efficiency are two key goals for new GPU architectures. A number of changes to the SM in the Maxwell architecture improved its efficiency compared to Kepler. Pascal builds on this and incorporates additional improvements that increase performance per watt even further over Maxwell. While TSMC’s 16nm Fin-FET manufacturing process plays an important role, many GPU architectural modifications were also implemented to further reduce power consumption while maintaining high performance.

 

The following table provides a high-level comparison of Tesla P100 specifications compared to previous-generation Tesla GPU accelerators.

Tesla Products Tesla K40 Tesla M40 Tesla P100
GPU GK110 (Kepler) GM200 (Maxwell) GP100 (Pascal)
SMs 15 24 56
TPCs 15 24 28
FP32 CUDA Cores / SM 192 128 64
FP32 CUDA Cores / GPU 2880 3072 3584
FP64 CUDA Cores / SM 64 4 32
FP64 CUDA Cores / GPU 960 96 1792
Base Clock 745 MHz 948 MHz 1328 MHz
GPU Boost Clock 810/875 MHz  1114 MHz 1480 MHz
FP64 GFLOPs 1680 213 5304[1]
Texture Units 240 192 224
Memory Interface 384-bit GDDR5 384-bit GDDR5 4096-bit HBM2
Memory Size Up to 12 GB Up to 24 GB 16 GB
L2 Cache Size 1536 KB 3072 KB 4096 KB
Register File Size / SM 256 KB 256 KB 256 KB
Register File Size / GPU 3840 KB 6144 KB 14336 KB
TDP 235 Watts 250 Watts 300 Watts
Transistors 7.1 billion 8 billion 15.3 billion
GPU Die Size 551 mm² 601 mm² 610 mm²
Manufacturing Process 28-nm 28-nm 16-nm

 

Block Diagrams: 

The Pascal GP100 Streaming Multiprocessor

 

 

Pascal GP100 Block Diagram

Source: NVIDIA (https://devblogs.nvidia.com/parallelforall/inside-pascal/)

Desktop: i5-4690K CPU | Asus Z97-A  Motherboard | 8 GB Dual Channel 1600MHz CL9 RAM | NZXT S340 White Case | Intel 535 Series 240GB SSD | CM V550 PSU | LG 25UM56-P Ultrawide Monitor | Cooler Master Hyper 212 EVO CPU Cooler | Logitech G602 Mouse | Sennheiser HD598 Headphones | Windows 10

Laptop: HP Pavilion dv6t-7000 Quad Edition | i7-3610qm | GT650M 1GB | 8GB DDR3 Dual Channel 1600MHz RAM | Crucial MX100 512GB SSD | 15.6" 1080 by 1920 | Windows 7 Home Premium

Link to comment
Share on other sites
Link to post
Share on other sites
Posted

May you please change the text to without format. So the white background isn't there? Us on the dark theme cannot read the text all that well. Thanks.

GPU: XFX RX 7900 XTX

CPU: Ryzen 7 7800X3D

Link to comment
Share on other sites
Link to post
Share on other sites
Posted

So they added more cores and Slapped DP Units and stuff in there... Anyone have a run down of what these do? 

Computing enthusiast. 
I use to be able to input a cheat code now I've got to input a credit card - Total Biscuit
 

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
4 minutes ago, Deletive said:

So they added more cores and Slapped DP Units and stuff in there... Anyone have a run down of what these do? 

Yup that is the only thing they have done. Just looking at the size of the card it just looks like some cores slapped on. Yup 100%

▓ƆԀ▓ S▓₱▓Ɇ▓c▓s▓: i7 6ʇɥפᴉƎ00K (4.4ghz), Asus DeLuxe X99A II, GT҉X҉1҉0҉8҉0 Zotac Amp ExTrꍟꎭe),Si6F4Gb D???????r PlatinUm, EVGA G2 Sǝʌǝᘉ5ᙣᙍᖇᓎᙎᗅᖶt, Phanteks Enthoo Primo, 3TB WD Black, 500gb 850 Evo, H100iGeeTeeX, Windows 10, K70 R̸̢̡̭͍͕̱̭̟̩̀̀̃́̃͒̈́̈́͑̑́̆͘͜ͅG̶̦̬͊́B̸͈̝̖͗̈́, G502, HyperX Cloud 2s, Asus MX34. פN∩SW∀S 960 EVO

Just keeping this here as a backup 9̵̨̢̨̧̧̡̧̡̧̡̧̡̡̢̢̡̢̧̡̢̡̡̢̧̛̛̛̛̛̛̱̖͈̠̝̯̹͉̝̞̩̠̹̺̰̺̲̳͈̞̻̜̫̹̱̗̣͙̻̘͎̲̝͙͍͔̯̲̟̞͚̖̘͉̭̰̣͎͕̼̼̜̼͕͎̣͇͓͓͎̼̺̯͈̤̝͖̩̭͍̣̱̞̬̺̯̼̤̲͎̖̠̟͍̘̭͔̟̗̙̗̗̤̦͍̫̬͔̦̳̗̳͔̞̼̝͍̝͈̻͇̭̠͈̳͍̫̮̥̭͍͔͈̠̹̼̬̰͈̤͚̖̯͍͉͖̥̹̺͕̲̥̤̺̹̹̪̺̺̭͕͓̟̳̹͍̖͎̣̫͓͍͈͕̳̹̙̰͉͙̝̜̠̥̝̲̮̬͕̰̹̳͕̰̲̣̯̫̮͙̹̮͙̮̝̣͇̺̺͇̺̺͈̳̜̣̙̻̣̜̻̦͚̹̩͓͚̖͍̥̟͍͎̦͙̫̜͔̭̥͈̬̝̺̩͙͙͉̻̰̬̗̣͖̦͎̥̜̬̹͓͈͙̤̜̗͔̩̖̳̫̑̀̂̽̈́̈́̿͒̿̋̊͌̾̄̄̒̌͐̽̿̊͑̑̆͗̈̎̄͒̑̋͛̑͑̂͑̀͐̀͑̓͊̇͆̿͑͛͛͆́͆̓̿̇̀̓͑͆͂̓̾̏͊̀̇̍̃́̒̎̀̒̄̓̒̐̑̊̏̌̽̓͂͋̓̐̓͊̌͋̀̐̇̌̓̔͊̈̇́̏͒̋͊̓̆̋̈̀̌̔͆͑̈̐̈̍̀̉̋̈́͊̽͂̿͌͊̆̾̉͐̿̓̄̾͑̈́͗͗̂̂́̇͂̀̈́́̽̈́̓̓͂̽̓̀̄͌̐̔̄̄͒͌̈́̅̉͊̂͒̀̈́̌͂̽̀̑̏̽̀͑̐̐͋̀̀͋̓̅͋͗̍́͗̈́̆̏̇͊̌̏̔̑̐̈́͑̎͑͆̏̎́̑̍̏̒̌̊͘͘̚̕̚̕̕̚̕̚̕̕͜͜͜͜͜͝͝͠͠͝͝͝͝͝͝͝͠͝͝ͅͅͅͅͅͅͅ8̵̨̛̛̛̛̮͍͕̥͉̦̥̱̞̜̫̘̤̖̬͍͇͓̜̻̪̤̣̣̹̑͑̏̈́̐̐́̎͒̔͒̌̑̓̆̓͑̉̈́́͋̌͋͐͛͋̃̍̽̊͗͋͊̂̅͊͑́͋͛̉̏̓͌̾̈́̀͛͊̾͑̌̀̀̌̓̏̑́̄̉̌͂́͛̋͊̄͐͊̈́̀̌̆̎̿̓̔̍̎̀̍̚̕̕͘͘͘̕̚͝͝͠͠͠0̶̡̡̡̢̨̨͕̠̠͉̺̻̯̱̘͇̥͎͖̯͕̖̬̭͔̪̪͎̺̠̤̬̬̤̣̭̣͍̥̱̘̳̣̤͚̭̥͚̦͙̱̦͕̼͖͙͕͇̭͓͉͎̹̣̣͕̜͍͖̳̭͕̼̳̖̩͍͔̱̙̠̝̺̰̦̱̿̄̀͐͜͜ͅͅt̶̡̨̡̨̧̢̧̢̨̧̧̧̧̢̡̨̨̢̨̢̧̢̛̛̛̛̛̠͍̞̮͇̪͉̩̗̗͖̫͉͎͓̮̣̘̫͔̘̬̮̙̯̣͕͓̲̣͓͓̣̹̟͈̱͚̘̼̙̖̖̼̙̜̝͙̣̠̪̲̞̖̠̯̖̠̜̱͉̲̺͙̤̻̦̜͎̙̳̺̭̪̱͓̦̹̺͙̫̖̖̰̣͈͍̜̺̘͕̬̥͇̗̖̺̣̲̫̟̣̜̭̟̱̳̳̖͖͇̹̯̜̹͙̻̥̙͉͕̜͎͕̦͕̱͖͉̜̹̱̦͔͎̲̦͔̖̘̫̻̹̮̗̮̜̰͇̰͔̱͙̞̠͍͉͕̳͍̰̠̗̠̯̜̩͓̭̺̦̲̲͖̯̩̲̣̠͉̦̬͓̠̜̲͍̘͇̳̳͔̼̣͚̙͙͚͕̙̘̣̠͍̟̪̝̲͇͚̦̖͕̰̟̪͖̳̲͉͙̰̭̼̩̟̝̣̝̬̳͎̙̱͒̃̈͊̔͒͗̐̄̌͐͆̍͂̃̈́̾͗̅̐͒̓̆͛̂̾͋̍͂̂̄̇̿̈͌̅̈́̃̾̔̇̇̾̀͊͋̋̌̄͌͆͆̎̓̈́̾̊͊̇̌̔̈́̈́̀̐͊̊̍͑̊̈̓͑̀́̅̀̑̈́̽̃̽͛̇́̐̓̀͆̔̈̀̍̏̆̓̆͒̋́̋̍́̂̉͛̓̓̂̋̎́̒̏̈͋̃̽͆̓̀̔͑̈́̓͌͑̅̽́̐̍̉̑̓̈́͌̋̈́͂̊́͆͂̇̈́̔̃͌̅̈́͌͛̑̐̓̔̈́̀͊͛̐̾͐̔̾̈̃̈̄͑̓̋̇̉̉̚̕̚͘̕̚̚̕̕͜͜͜͜͜͜͜͜͜͜͜͜͜͝͝͝͠͝͝͝͝͝͠ͅͅͅͅͅi̵̢̧̢̧̡̧̢̢̧̢̢̢̡̡̡̧̧̡̡̧̛̛͈̺̲̫͕̞͓̥̖̭̜̫͉̻̗̭̖͔̮̠͇̩̹̱͈̗̭͈̤̠̮͙͇̲͙̰̳̹̲͙̜̟͚͎͓̦̫͚̻̟̰̣̲̺̦̫͓̖̯̝̬͉̯͓͈̫̭̜̱̞̹̪͔̤̜͙͓̗̗̻̟͎͇̺̘̯̲̝̫͚̰̹̫̗̳̣͙̮̱̲͕̺̠͉̫̖̟͖̦͉̟͈̭̣̹̱̖̗̺̘̦̠̯̲͔̘̱̣͙̩̻̰̠͓͙̰̺̠̖̟̗̖͉̞̣̥̝̤̫̫̜͕̻͉̺͚̣̝̥͇̭͎̖̦̙̲͈̲̠̹̼͎͕̩͓̖̥̘̱̜͙̹̝͔̭̣̮̗̞̩̣̬̯̜̻̯̩̮̩̹̻̯̬̖͂̈͂̒̇͗͑̐̌̎̑̽̑̈̈́͑̽́̊͋̿͊͋̅̐̈́͑̇̿̈́̌͌̊̅͂̎͆̏̓͂̈̿̏̃͑̏̓͆̔̋̎̕͘͘͘͜͜͜͜͜͜͜͝͝͠͠ͅͅͅͅͅͅͅͅͅZ̴̧̢̨̢̧̢̢̡̧̢̢̢̨̨̨̡̨̧̢̧̛̛̬̖͈̮̝̭̖͖̗̹̣̼̼̘̘̫̠̭̞͙͔͙̜̠̗̪̠̼̫̻͓̳̟̲̳̻̙̼͇̺͎̘̹̼͔̺̹̬̯̤̮̟͈̭̻͚̣̲͔͙̥͕̣̻̰͈̼̱̺̤̤͉̙̦̩̗͎̞͓̭̞̗͉̳̭̭̺̹̹̮͕̘̪̞̱̥͈̹̳͇̟̹̱̙͚̯̮̳̤͍̪̞̦̳̦͍̲̥̳͇̪̬̰̠͙͕̖̝̫̩̯̱̘͓͎̪͈̤̜͎̱̹̹̱̲̻͎̖̳͚̭̪̦̗̬͍̯̘̣̩̬͖̝̹̣̗̭͖̜͕̼̼̲̭͕͔̩͓̞̝͓͍̗̙̯͔̯̞̝̳̜̜͉̖̩͇̩̘̪̥̱͓̭͎͖̱̙̩̜͎̙͉̟͎͔̝̥͕͍͓̹̮̦̫͚̠̯͓̱͖͔͓̤͉̠͙̋͐̀͌̈́͆̾͆̑̔͂͒̀̊̀͋͑̂͊̅͐̿́̈́̐̀̏̋̃̄͆͒̈́̿̎́́̈̀̀͌̔͋͊̊̉̿͗͊͑̔͐̇͆͛̂̐͊̉̄̈́̄̐͂͂͒͑͗̓͑̓̾̑͋̒͐͑̾͂̎̋̃̽̂̅̇̿̍̈́́̄̍͂͑̏̐̾̎̆̉̾͂̽̈̆̔́͋͗̓̑̕͘̕͘͜͜͜͜͜͝͝͝͝͠͠͝ͅo̶̪͆́̀͂̂́̄̅͂̿͛̈́̿͊͗́͘͝t̴̡̨̧̨̧̡̧̨̡̢̧̢̡̨̛̪͈̣̭̺̱̪̹̺̣̬̖̣̻͈̞̙͇̩̻̫͈̝̭̟͎̻̟̻̝̱͔̝̼͍̞̼̣̘̤̯͓͉̖̠̤͔̜̙͚͓̻͓̬͓̻̜̯̱̖̳̱̗̠̝̥̩͓̗̪̙͓̖̠͎̗͎̱̮̯̮͙̩̫̹̹̖͙̙͖̻͈̙̻͇͔̙̣̱͔̜̣̭̱͈͕̠̹͙̹͇̻̼͎͍̥̘͙̘̤̜͎̟͖̹̦̺̤͍̣̼̻̱̲͎̗̹͉͙̪̞̻̹͚̰̻͈͈͊̈́̽̀̎̃̊́̈́̏̃̍̉̇̑̂̇̏̀͊̑̓͛̽͋̈́͆́̊͊̍͌̈́̓͊̌̿̂̾̐͑̓̀́͒̃̋̓͆̇̀͊̆͗̂͑͐̀͗̅̆͘̕͘̕̕͜͜͝͝͝͝͝͝͝ͅͅͅͅͅͅͅͅͅḁ̶̢̡̨̧̡̡̨̨̧̨̡̡̢̧̨̡̡̛̛̛͍̱̳͚͕̩͍̺̪̻̫̙͈̬͙̖͙̬͍̬̟̣̝̲̼̜̼̺͎̥̮̝͙̪̘̙̻͖͇͚͙̣̬̖̲̲̥̯̦̗̰̙̗̪̞̗̩̻̪̤̣̜̳̩̦̻͓̞̙͍͙̫̩̹̥͚̻̦̗̰̲̙̫̬̱̺̞̟̻͓̞͚̦̘̝̤͎̤̜̜̥̗̱͈̣̻̰̮̼̙͚͚̠͚̲̤͔̰̭̙̳͍̭͎̙͚͍̟̺͎̝͓̹̰̟͈͈̖̺͙̩̯͔̙̭̟̞̟̼̮̦̜̳͕̞̼͈̜͍̮͕̜͚̝̦̞̥̜̥̗̠̦͇͖̳͈̜̮̣͚̲̟͙̎̈́́͊̔̑̽̅͐͐͆̀͐́̓̅̈͑͑̍̿̏́͆͌̋̌̃̒̽̀̋̀̃̏̌́͂̿̃̎̐͊̒̀̊̅͒̎͆̿̈́̑̐̒̀̈́̓̾͋͆̇̋͒̎̈̄̓̂͊̆͂̈́̒̎͐̇̍̆̋̅̿̔͒̄̇̂̋̈́͆̎̔̇͊̊̈́̔̏͋́̀͂̈́̊͋͂̍̾̓͛̇̔̚͘̚̕̚͘͘̕̕̕̚͘͘̚̕̚̕͜͜͜͝͝͝͝͝͝͝͝ͅͅͅͅͅç̵̧̢̨̢̢̢̧̧̡̨̡̢̧̧̧̨̡̡̨̨̢̢̢̧̨̢̨̢̛̛͉̗̠͇̹̖̝͕͚͎̟̻͓̳̰̻̺̞̣͚̤͙͍͇̗̼͖͔͕͙͖̺͙̖̹̘̘̺͓̜͍̣̰̗̖̺̗̪̘̯̘͚̲͚̲̬̞̹̹͕̭͔̳̘̝̬͉̗̪͉͕̞̫͔̭̭̜͉͔̬̫͙̖̙͚͔͙͚͍̲̘͚̪̗̞̣̞̲͎͔͖̺͍͎̝͎͍̣͍̩̟͈͕̗͉̪̯͉͎͖͍̖͎̖̯̲̘̦̟̭͍͚͓͈͙̬͖̘̱̝̜̘̹̩̝̥̜͎̬͓̬͙͍͇͚̟̫͇̬̲̥̘̞̘̟̘̝̫͈̙̻͇͎̣̪̪̠̲͓͉͙͚̭̪͇̯̠̯̠͖̞̜͓̲͎͇̼̱̦͍͉͈͕͉̗̟̖̗̱̭͚͎̘͓̬͍̱͍̖̯̜̗̹̰̲̩̪͍̞̜̫̩̠͔̻̫͍͇͕̰̰̘͚͈̠̻̮͊̐̿̏̐̀̇̑̐̈͛͑͑̍̑̔̃̈́̓̈́̇̐͑̐̊̆͂̀̏͛̊̔̍̽͗͋̊̍̓̈́̏̅͌̀̽́̑͒͒̓͗̈́̎͌͂̕̚͘͘͜͜͜͜͜͠͝͝͝͝ͅͅͅͅͅͅͅS̵̡̡̧̧̨̨̡̢̡̡̡̡̧̧̡̧̢̫̯͔̼̲͉͙̱̮̭̗͖̯̤͙̜͚̰̮̝͚̥̜̞̠̤̺̝͇̻̱͙̩̲̺͍̳̤̺̖̝̳̪̻̗̮̪̖̺̹̭͍͇̗̝̱̻̳̝̖̝͎̙͉̞̱̯̙̜͇̯̻̞̱̭̗͉̰̮̞͍̫̺͙͎̙̞̯̟͓͉̹̲͖͎̼̫̩̱͇̲͓̪͉̺̞̻͎̤̥̭̺̘̻̥͇̤̖̰̘̭̳̫̙̤̻͇̪̦̭̱͎̥̟͖͕̣̤̩̟̲̭̹̦̹̣͖̖͒̈́̈́̓͗̈̄͂̈́̅̐̐̿̎̂͗̎̿̕͘͜͜͜͜͝͝ͅͅt̸̡̡̧̧̨̡̢̛̥̥̭͍̗͈̩͕͔͔̞̟͍̭͇̙̺̤͚͎͈͎͕̱͈̦͍͔͓̬͚̗̰̦͓̭̰̭̎̀̂̈́̓̒̈́̈́̂̄̋́̇̂͐͒̋̋̉͐̉̏̇͋̓̈́͐̾͋̒͒͐̊̊̀̄͆̄͆̑͆̇̊̓̚̚̕̚̕͜͠͝͝ͅͅơ̵̡̨̡̡̡̨̛̺͕̼͔̼̪̳͖͓̠̘̘̳̼͚͙͙͚̰͚͚͖̥̦̥̘̖̜̰͔̠͕̦͎̞̮͚͕͍̤̠̦͍̥̝̰̖̳̫̮̪͇̤̱̜͙͔̯͙̙̼͇̹̥̜͈̲̺̝̻̮̬̼̫̞̗̣̪̱͓̺̜̠͇͚͓̳̹̥̳̠͍̫͈̟͈̘̯̬̞͔̝͍͍̥̒̐͗͒͂͆̑̀̿̏́̀͑͗̐́̀̾̓́̌̇̒̈́̌̓͐̃̈́̒̂̀̾͂̊̀̂͐̃̄̓̔̽̒̈́̇̓͌̇̂̆̒̏̊̋͊͛͌̊̇̒̅͌̄̎̔̈́͊́̽̋̈̇̈́́͊̅͂̎̃͌͊͛͂̄̽̈́̿͐̉̽̿́́̉͆̈́̒́̂̾̄̇̌̒̈̅̍̿̐͑̓͊̈́̈̋̈́̉̍̋̊̈̀̈́̾̿̌̀̈́͌̑̍́̋̒̀̂̈́́̾̏̐̅̈̑͗͐̈͂̄̾̄̈́̍̉͑͛͗͋̈́̃̄̊́́͐̀̀̽̇̓̄̓̃͋͋̂̽̔̀̎͌̈́̈́̑̓̔̀̓͐͛͆̿̋͑͛̈́͂̅̋̅͆͗̇́̀̒́̏͒̐̍͂̓͐͐̇̂̉̑̊͑̉̋̍͊̄̀͂̎͒̔͊̃̏̕̚̕̕͘͘͘̚͘̚͘̕͘̚͘̚̚̚̕͘͜͜͜͝͝͠͠͝͝͠͠͝͝͝͝͝͝͝͝͝ͅͅͅc̴̨̡̢̢̢̡̡̢̛̛̛̻͇̝̣͉͚͎͕̻̦͖̤̖͇̪̩̤̻̭̮̙̰̖̰̳̪̱̹̳̬͖̣͙̼̙̰̻̘͇͚̺̗̩̫̞̳̼̤͔͍͉̟͕̯̺͈̤̰̹̍̋́͆̾̆̊͆͋̀͑͒̄̿̄̀̂͋̊͆́͑̑̽͊̓́̔̽̌͊̄͑͒͐̑͗̿̃̀̓̅́̿͗̈́͌̋̀̏̂͌̓́̇̀͒͋̌̌̅͋͌̆͐̀̔̒͐̊̇̿̽̀̈́̃̒̋̀̈́̃̏̂̊͗̑̊̈̇̀̌͐̈́̉̂̏͊̄͐̈̽͒̏̒̓́̌̓̅́̓̃͐͊͒̄͑̒͌̍̈́̕͘̚͘̕͘̚̕͜͝͠͝͝͝ͅǩ̴̢̢̢̧̨̢̢̢̨̨̨̢̢̢̨̧̨̡̡̢̛̛̛̛̛̛̛̜̥̩̙͕̮̪̻͈̘̯̼̰̜͚̰͖̬̳͖̣̭̼͔̲͉̭̺͚̺̟͉̝̱̲͎͉̙̥̤͚͙̬̪̜̺͙͍̱̞̭̬̩̖̤̹̤̺̦͈̰̗̰͍͇̱̤̬̬͙̙̲̙̜͖͓̙̟̙̯̪͍̺̥͔͕̝̳̹̻͇̠̣͈̰̦͓͕̩͇͈͇̖͙͍̰̲̤̞͎̟̝̝͈͖͔͖̦̮̗̬̞̞̜̬̠̹̣̣̲̮̞̤̜̤̲̙͔͕̯͔͍̤͕̣͔͙̪̫̝̣̰̬̬̭̞͔̦̟̥̣̻͉͈̮̥̦̮̦͕̤͇̺͆͆̈͗̄̀̌̔̈́̈̉̾̊̐̆̂͛̀̋́̏̀̿͒̓̈́̈́͂̽̾͗͊̋̐̓̓̀̃̊̊͑̓̈̎̇͑̆̂̉̾̾̑͊̉̃́̑͌̀̌̐̅̃̿̆̎̈́̀̒́͛̓̀̊́̋͛͒͊̆̀̃̊͋̋̾̇̒̋͂̏͗͆̂̔́̐̀́͗̅̈̋̂̎̒͊̌̉̈̈́͌̈́̔̾̊̎́͐͒̋̽̽́̾̿̚̕͘͘̚̕̕̕̚̚̕̚̕͘͜͜͜͝͠͝͝͝͝͝͝͝͝ͅͅͅͅͅͅB̸̢̧̨̡̢̧̨̡̡̨̡̨̡̡̡̢̨̢̨̛̛̛̛̛̛͉̞͚̰̭̲͈͎͕͈̦͍͈̮̪̤̻̻͉̫̱͔̞̫̦̰͈̗̯̜̩̪̲̻̖̳͖̦͎͔̮̺̬̬̼̦̠̪̤͙͍͓̜̥̙̖̫̻̜͍̻̙̖̜̹͔̗̪̜̖̼̞̣̠̫͉̯̮̤͈͎̝̪͎͇͙̦̥͙̳̫̰̪̣̱̘̤̭̱͍̦͔̖͎̺̝̰̦̱̣͙̙̤͚̲͔̘̱̜̻͔̥̻͖̭͔̜͉̺͕͙͖̜͉͕̤͚̠̩̮̟͚̗͈͙̟̞̮̬̺̻̞͔̥͉͍̦̤͓̦̻̦̯̟̰̭̝̘̩̖̝͔̳͉̗̖̱̩̩̟͙͙͛̀͐̈́̂̇͛̅̒̉̏̈́̿͐́̏̃̏̓̌̽͐̈́͛̍͗͆͛̋̔̉͂̔̂̓̌͌͋̂͆̉͑̊̎́̈́̈̂͆͑́̃̍̇̿̅̾́́̿̅̾̆̅̈́̈̓͒͌͛̃͆̋͂̏̓̅̀͂̽̂̈̈́̎̾̐͋͑̅̍̈́̑̅̄͆̓̾̈́͐̎̊͐̌̌̓͊̊̔̈́̃͗̓͊͐̌͆̓͗̓̓̾̂̽͊͗́́́̽͊͆͋͊̀̑̿̔͒̏̈́́̏͆̈́͋̒͗͂̄̇̒͐̃͑̅̍͒̎̈́̌̋́̓͂̀̇͛̋͊͆̈́̋́̍̃͒̆̕̚̚̕̕̕͘̕̚̚͘̕͜͜͜͜͝͠͠͝͠͝͝͝͝͠͝͝͝͝ͅͅͅͅͅI̵̡̢̧̨̡̢̨̡̡̢̡̧̡̢̢̢̡̢̛̛͕͎͕̩̠̹̩̺̣̳̱͈̻̮̺̟̘̩̻̫͖̟͓̩̜̙͓͇̙̱̭̰̻̫̥̗̠͍͍͚̞̘̫͉̬̫̖̖̦͖͉̖̩̩̖̤̺̥̻̝͈͎̻͓̟̹͍̲͚͙̹̟̟̯͚̳̟͕̮̻̟͈͇̩̝̼̭̯͚͕̬͇̲̲̯̰̖̙̣̝͇̠̞̙͖͎̮̬̳̥̣̺̰͔̳̳̝̩̤̦̳̞̰̩̫̟͚̱̪̘͕̫̼͉̹̹̟̮̱̤̜͚̝̠̤̖̮̯̳͖̗̹̞̜̹̭̿̏͋̒͆̔̄̃̾̓͛̾̌́̅̂͆̔͌͆͋̔̾́̈̇̐̄̑̓̂̾́̄̿̓̅̆͌̉̎̏̄͛̉͆̓̎͒͘̕̕͜͜͜͜͜͜͜͝͠ͅͅƠ̷̢̛̛̛̛̛̛̛̛̟̰͔͔͇̲̰̮̘̭̭̖̥̟̘̠̬̺̪͇̲͋͂̅̈́̍͂̽͗̾͒̇̇̒͐̍̽͊́̑̇̑̾̉̓̈̾͒̍̌̅̒̾̈́̆͌̌̾̎̽̐̅̏́̈̔͛̀̋̃͊̒̓͗͒̑͒̃͂̌̄̇̑̇͛̆̾͛̒̇̍̒̓̀̈́̄̐͂̍͊͗̎̔͌͛̂̏̉̊̎͗͊͒̂̈̽̊́̔̊̃͑̈́̑̌̋̓̅̔́́͒̄̈́̈̂͐̈̅̈̓͌̓͊́̆͌̉͐̊̉͛̓̏̓̅̈́͂̉̒̇̉̆̀̍̄̇͆͛̏̉̑̃̓͂́͋̃̆̒͋̓͊̄́̓̕̕̕̚͘͘͘̚̕̚͘̕̕͜͜͝͝͝͠͝͝͝͝͠ͅS̷̢̨̧̢̡̨̢̨̢̨̧̧̨̧͚̱̪͇̱̮̪̮̦̝͖̜͙̘̪̘̟̱͇͎̻̪͚̩͍̠̹̮͚̦̝̤͖̙͔͚̙̺̩̥̻͈̺̦͕͈̹̳̖͓̜͚̜̭͉͇͖̟͔͕̹̯̬͍̱̫̮͓̙͇̗̙̼͚̪͇̦̗̜̼̠͈̩̠͉͉̘̱̯̪̟͕̘͖̝͇̼͕̳̻̜͖̜͇̣̠̹̬̗̝͓̖͚̺̫͛̉̅̐̕͘͜͜͜͜ͅͅͅ.̶̨̢̢̨̢̨̢̛̻͙̜̼̮̝̙̣̘̗̪̜̬̳̫̙̮̣̹̥̲̥͇͈̮̟͉̰̮̪̲̗̳̰̫̙͍̦̘̠̗̥̮̹̤̼̼̩͕͉͕͇͙̯̫̩̦̟̦̹͈͔̱̝͈̤͓̻̟̮̱͖̟̹̝͉̰͊̓̏̇͂̅̀̌͑̿͆̿̿͗̽̌̈́̉̂̀̒̊̿͆̃̄͑͆̃̇͒̀͐̍̅̃̍̈́̃̕͘͜͜͝͠͠z̴̢̢̡̧̢̢̧̢̨̡̨̛̛̛̛̛̛̛̛̲͚̠̜̮̠̜̞̤̺͈̘͍̻̫͖̣̥̗̙̳͓͙̫̫͖͍͇̬̲̳̭̘̮̤̬̖̼͎̬̯̼̮͔̭̠͎͓̼̖̟͈͓̦̩̦̳̙̮̗̮̩͙͓̮̰̜͎̺̞̝̪͎̯̜͈͇̪̙͎̩͖̭̟͎̲̩͔͓͈͌́̿͐̍̓͗͑̒̈́̎͂̋͂̀͂̑͂͊͆̍͛̄̃͌͗̌́̈̊́́̅͗̉͛͌͋̂̋̇̅̔̇͊͑͆̐̇͊͋̄̈́͆̍̋̏͑̓̈́̏̀͒̂̔̄̅̇̌̀̈́̿̽̋͐̾̆͆͆̈̌̿̈́̎͌̊̓̒͐̾̇̈́̍͛̅͌̽́̏͆̉́̉̓̅́͂͛̄̆͌̈́̇͐̒̿̾͌͊͗̀͑̃̊̓̈̈́̊͒̒̏̿́͑̄̑͋̀̽̀̔̀̎̄͑̌̔́̉̐͛̓̐̅́̒̎̈͆̀̍̾̀͂̄̈́̈́̈́̑̏̈́̐̽̐́̏̂̐̔̓̉̈́͂̕̚̕͘͘̚͘̚̕̚̚̚͘̕̕̕͜͜͝͠͠͝͝͝͝͠͝͝͝͠͝͝͝͝͝͝ͅͅͅī̸̧̧̧̡̨̨̢̨̛̛̘͓̼̰̰̮̗̰͚̙̥̣͍̦̺͈̣̻͇̱͔̰͈͓͖͈̻̲̫̪̲͈̜̲̬̖̻̰̦̰͙̤̘̝̦̟͈̭̱̮̠͍̖̲͉̫͔͖͔͈̻̖̝͎̖͕͔̣͈̤̗̱̀̅̃̈́͌̿̏͋̊̇̂̀̀̒̉̄̈́͋͌̽́̈́̓̑̈̀̍͗͜͜͠͠ͅp̴̢̢̧̨̡̡̨̢̨̢̢̢̨̡̛̛͕̩͕̟̫̝͈̖̟̣̲̖̭̙͇̟̗͖͎̹͇̘̰̗̝̹̤̺͉͎̙̝̟͙͚̦͚͖̜̫̰͖̼̤̥̤̹̖͉͚̺̥̮̮̫͖͍̼̰̭̤̲͔̩̯̣͖̻͇̞̳̬͉̣̖̥̣͓̤͔̪̙͎̰̬͚̣̭̞̬͎̼͉͓̮͙͕̗̦̞̥̮̘̻͎̭̼͚͎͈͇̥̗͖̫̮̤̦͙̭͎̝͖̣̰̱̩͎̩͎̘͇̟̠̱̬͈̗͍̦̘̱̰̤̱̘̫̫̮̥͕͉̥̜̯͖̖͍̮̼̲͓̤̮͈̤͓̭̝̟̲̲̳̟̠͉̙̻͕͙̞͔̖͈̱̞͓͔̬̮͎̙̭͎̩̟̖͚̆͐̅͆̿͐̄̓̀̇̂̊̃̂̄̊̀͐̍̌̅͌̆͊̆̓́̄́̃̆͗͊́̓̀͑͐̐̇͐̍́̓̈́̓̑̈̈́̽͂́̑͒͐͋̊͊̇̇̆̑̃̈́̎͛̎̓͊͛̐̾́̀͌̐̈́͛̃̂̈̿̽̇̋̍͒̍͗̈͘̚̚͘̚͘͘͜͜͜͜͜͜͠͠͝͝ͅͅͅ☻♥■∞{╚mYÄÜXτ╕○\╚Θº£¥ΘBM@Q05♠{{↨↨▬§¶‼↕◄►☼1♦  wumbo╚̯̪̣͕̙̩̦͓͚̙̱̘̝̏̆ͤ̊̅ͩ̓̏̿͆̌Θ̼̯͉ͭͦ̃͊͑̉ͯͤ̈́ͬ͐̈́͊ͤͅº͍̪͇͖̝̣̪̙̫̞̦̥ͨ̂ͧ̄̿£̺̻̹̠̯͙͇̳ͬ̃̿͑͊ͨͣ╚̯̪̣͕̙̩̦͓͚̙̱̘̝̏̆ͤ̊̅ͩ̓̏̿͆̌Θ̼̯͉ͭͦ̃͊͑̉ͯͤ̈́ͬ͐̈́͊ͤͅº͍̪͇͖̝̣̪̙̫̞̦̥ͨ̂ͧ̄̿£̺̻̹̠̯͙͇̳ͬ̃̿͑͊ͨͣ╚̯̪̣͕̙̩̦͓͚̙̱̘̝̏̆ͤ̊̅ͩ̓̏̿͆̌Θ̼̯͉ͭͦ̃͊͑̉ͯͤ̈́ͬ͐̈́͊ͤͅº͍̪͇͖̝̣̪̙̫̞̦̥ͨ̂ͧ̄̿£̺̻̹̠̯͙͇̳ͬ̃̿͑͊ͨͣ╚̯̪̣͕̙̩̦͓͚̙̱̘̝̏̆ͤ̊̅ͩ̓̏̿͆̌Θ̼̯͉ͭͦ̃͊͑̉ͯͤ̈́ͬ͐̈́͊ͤͅº͍̪͇͖̝̣̪̙̫̞̦̥ͨ̂ͧ̄̿£̺̻̹̠̯͙͇̳ͬ̃̿͑͊ͨͣ╚̯̪̣͕̙̩̦͓͚̙̱̘̝̏̆ͤ̊̅ͩ̓̏̿͆̌Θ̼̯͉ͭͦ̃͊͑̉ͯͤ̈́ͬ͐̈́͊ͤͅº͍̪͇͖̝̣̪̙̫̞̦̥ͨ̂ͧ̄̿£̺̻̹̠̯͙͇̳ͬ̃̿͑͊ͨͣ╚̯̪̣͕̙̩̦͓͚̙̱̘̝̏̆ͤ̊̅ͩ̓̏̿͆̌Θ̼̯͉ͭͦ̃͊͑̉ͯͤ̈́ͬ͐̈́͊ͤͅº͍̪͇͖̝̣̪̙̫̞̦̥ͨ̂ͧ̄̿£̺̻̹̠̯͙͇̳ͬ̃̿͑͊ͨͣ╚̯̪̣͕̙̩̦͓͚̙̱̘̝̏̆ͤ̊̅ͩ̓̏̿͆̌Θ̼̯͉ͭͦ̃͊͑̉ͯͤ̈́ͬ͐̈́͊ͤͅº͍̪͇͖̝̣̪̙̫̞̦̥ͨ̂ͧ̄̿£̺̻̹̠̯͙͇̳ͬ̃̿͑͊ͨͣ╚̯̪̣͕̙̩̦͓͚̙̱̘̝̏̆ͤ̊̅ͩ̓̏̿͆̌Θ̼̯͉ͭͦ̃͊͑̉ͯͤ̈́ͬ͐̈́͊ͤͅº͍̪͇͖̝̣̪̙̫̞̦̥ͨ̂ͧ̄̿£̺̻̹̠̯͙͇̳ͬ̃̿͑͊ͨͣ╚̯̪̣͕̙̩̦͓͚̙̱̘̝̏̆ͤ̊̅ͩ̓̏̿͆̌Θ̼̯͉ͭͦ̃͊͑̉ͯͤ̈́ͬ͐̈́͊ͤͅº͍̪͇͖̝̣̪̙̫̞̦̥ͨ̂ͧ̄̿£̺̻̹̠̯͙͇̳ͬ̃̿͑͊ͨͣ╚̯̪̣͕̙̩̦͓͚̙̱̘̝̏̆ͤ̊̅ͩ̓̏̿͆̌Θ̼̯͉ͭͦ̃͊͑̉ͯͤ̈́ͬ͐̈́͊ͤͅº͍̪͇͖̝̣̪̙̫̞̦̥ͨ̂ͧ̄̿£̺̻̹̠̯͙͇̳ͬ̃̿͑͊ͨͣ╚̯̪̣͕̙̩̦͓͚̙̱̘̝̏̆ͤ̊̅ͩ̓̏̿͆̌Θ̼̯͉ͭͦ̃͊͑̉ͯͤ̈́ͬ͐̈́͊ͤͅº͍̪͇͖̝̣̪̙̫̞̦̥ͨ̂ͧ̄̿£̺̻̹̠̯͙͇̳ͬ̃̿͑͊ͨͣ╚̯̪̣͕̙̩̦͓͚̙̱̘̝̏̆ͤ̊̅ͩ̓̏̿͆̌Θ̼̯͉ͭͦ̃͊͑̉ͯͤ̈́ͬ͐̈́͊ͤͅº͍̪͇͖̝̣̪̙̫̞̦̥ͨ̂ͧ̄̿£̺̻̹̠̯͙͇̳ͬ̃̿͑͊ͨͣ╚̯̪̣͕̙̩̦͓͚̙̱̘̝̏̆ͤ̊̅ͩ̓̏̿͆̌Θ̼̯͉ͭͦ̃͊͑̉ͯͤ̈́ͬ͐̈́͊ͤͅº͍̪͇͖̝̣̪̙̫̞̦̥ͨ̂ͧ̄̿£̺̻̹̠̯͙͇̳ͬ̃̿͑͊ͨͣ╚̯̪̣͕̙̩̦͓͚̙̱̘̝̏̆ͤ̊̅ͩ̓̏̿͆̌Θ̼̯͉ͭͦ̃͊͑̉ͯͤ̈́ͬ͐̈́͊ͤͅº͍̪͇͖̝̣̪̙̫̞̦̥ͨ̂ͧ̄̿£̺̻̹̠̯͙͇̳ͬ̃̿͑͊ͨͣ╚̯̪̣͕̙̩̦͓͚̙̱̘̝̏̆ͤ̊̅ͩ̓̏̿͆̌Θ̼̯͉ͭͦ̃͊͑̉ͯͤ̈́ͬ͐̈́͊ͤͅº͍̪͇͖̝̣̪̙̫̞̦̥ͨ̂ͧ̄̿£̺̻̹̠̯͙͇̳ͬ̃̿͑͊ͨͣ

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
Just now, SamStrecker said:

Yup that is the only thing they have done. Just looking at the size of the card it just looks like some cores slapped on. Yup 100%

Given Pascal was the original design and Maxwell had to tear parts out to fit, this really shouldn't surprise us. That said, they've probably done much more than just increase the total number of clusters.

Software Engineer for Suncorp (Australia), Computer Tech Enthusiast, Miami University Graduate, Nerd

Link to comment
Share on other sites
Link to post
Share on other sites
Posted

Now to wait for AMD's upcoming GPUs

I wonder how 16nm would compare to 14nm

hello!

is it me you're looking for?

ᴾC SᴾeCS ᴰoWᴺ ᴮEᴸoW

Spoiler

Desktop: X99-PC

CPU: i7 5820k

Mobo: X99 Deluxe

Cooler: Dark Rock Pro 3

RAM: 32GB DDR4
GPU: GTX 1080

Storage: 1TB 850 Evo, 1TB HDD, bunch of external hard drives
PSU: EVGA G2 750w

Peripherals: Logitech G502, Ducky One 711

Audio: Xonar U7, O2 amplifier (RIP), HD6XX

Monitors: 4k 24" Dell monitor, 1080p 24" Asus monitor

 

Laptop:

-Overkill Dell XPS

Fully maxed out early 2017 Dell XPS 15, GTX 1050 4GB, 7700HQ, 1TB nvme SSD, 32GB RAM, 4k display. 97Whr battery :x 
Dell was having a $600 off sale for the fully specced out model, so I decided to get it :P

 

-Crapbook

Fully specced out early 2013 Macbook "pro" with gt 650m and constant 105c temperature on the CPU (GPU is 80-90C) when doing anything intensive...

A 2013 laptop with a regular sized battery still has better battery life than a 2017 laptop with a massive battery! I think this is a testament to apple's ability at making laptops, or maybe how little CPU technology has improved even 4+ years later (at least, until the recent introduction of 15W 4 core CPUs). Anyway, I'm never going to get a 35W CPU laptop again unless battery technology becomes ~5x better than as it is in 2018.

Apple knows how to make proper consumer-grade laptops (they don't know how to make pro laptops though). I guess this mostly software power efficiency related, but getting a mac makes perfect sense if you want a portable/powerful laptop that can do anything you want it to with great battery life.

 

 

Link to comment
Share on other sites
Link to post
Share on other sites
Posted

The most striking thing is that huge L2 cache.

Software Engineer for Suncorp (Australia), Computer Tech Enthusiast, Miami University Graduate, Nerd

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
4 minutes ago, SamStrecker said:

Yup that is the only thing they have done. Just looking at the size of the card it just looks like some cores slapped on. Yup 100%

Well I can notice a difference but I'm wondering what most of these difference make in performance and why.

2 minutes ago, patrickjp93 said:

The most striking thing is that huge L2 cache.

Wasn't L2 Cache on Maxwell already pretty big? 

Computing enthusiast. 
I use to be able to input a cheat code now I've got to input a credit card - Total Biscuit
 

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
Just now, Deletive said:

Well I can notice a difference but I'm wondering what most of these difference make in performance and why.

Wasn't L2 Cache on Maxwell already pretty big? 

2.5MB iirc.

Software Engineer for Suncorp (Australia), Computer Tech Enthusiast, Miami University Graduate, Nerd

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
4 minutes ago, Deletive said:

Well I can notice a difference but I'm wondering what most of these difference make in performance and why.

Wasn't L2 Cache on Maxwell already pretty big? 

 

3 minutes ago, patrickjp93 said:

2.5MB iirc.

2MB on GM204, 3MB on GM200

Forum Rules   |   Guide to Display Cables / Adapters

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
Just now, Glenwing said:

 

2MB on GM204, 3MB on GM200

Thank you kindly.

Software Engineer for Suncorp (Australia), Computer Tech Enthusiast, Miami University Graduate, Nerd

Link to comment
Share on other sites
Link to post
Share on other sites
Posted

Well I'm ready to upgrade my 680s. They have been struggling on my UW screen.

Main Rig "Rocinante" - Ryzen 9 5900X, EVGA FTW3 RTX 3080 Ultra Gaming, 32GB 3600MHz DDR4

Link to comment
Share on other sites
Link to post
Share on other sites
Posted

i might be missing something here, im dont know the nvidia lingo and normally skip this kind of posts as i know nothing of what this means...

 

but isnt the Tesla gpu for workstation and supercomputeres? so not the average consumer?

CPU: 6700k 4.6Ghz GPU: MSI GTX 1070 Gaming X MB: MSI Gaming M5 PSU: Evga 750 G2 Case: Phanteks EVOLV 

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
1 hour ago, Spectrez said:

i might be missing something here, im dont know the nvidia lingo and normally skip this kind of posts as i know nothing of what this means...

 

but isnt the Tesla gpu for workstation and supercomputeres? so not the average consumer?

so you DO know nvidia lingo.

liar.

Link to comment
Share on other sites
Link to post
Share on other sites
Posted

At least i have some more time to sell my old GTX 980 

Current Rig:   CPU: AMD 1950X @4Ghz. Cooler: Enermax Liqtech TR4 360. Motherboard:Asus Zenith Extreme. RAM: 8GB Crucial DDR4 3666. GPU: Reference GTX 970  SSD: 250GB Samsung 970 EVO.  HDD: Seagate Barracuda 7200.14 2TB. Case: Phanteks Enthoo Pro. PSU: Corsair RM1000X. OS: Windows 10 Pro UEFI mode  (installed on SSD)

Peripherals:  Display: Acer XB272 1080p 240Hz G Sync Keyboard: Corsair K95 RGB Brown Mouse: Logitech G502 RGB Headhet: Roccat XTD 5.1 analogue

Daily Devices:Sony Xperia XZ1 Compact and 128GB iPad Pro

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
8 hours ago, Deletive said:

So they added more cores and Slapped DP Units and stuff in there... Anyone have a run down of what these do? 

Not only that. If my understanding is correct they shrunk the size of their SMs to half the size of Maxwell SMs. So even better power efficiency. If you have smaller SMs then you can more finely tune the SMs so that you only use the ones you need.  Thus saving boat loads of power.

Judge a product on its own merits AND the company that made it.

How to setup MSI Afterburner OSD | How to make your AMD Radeon GPU more efficient with Radeon Chill | (Probably) Why LMG Merch shipping to the EU is expensive

Oneplus 6 (Early 2023 to present) | HP Envy 15" x360 R7 5700U (Mid 2021 to present) | Steam Deck (Late 2022 to present)

 

Mid 2023 AlTech Desktop Refresh - AMD R7 5800X (Mid 2023), XFX Radeon RX 6700XT MBA (Mid 2021), MSI X370 Gaming Pro Carbon (Early 2018), 32GB DDR4-3200 (16GB x2) (Mid 2022

Noctua NH-D15 (Early 2021), Corsair MP510 1.92TB NVMe SSD (Mid 2020), beQuiet Pure Wings 2 140mm x2 & 120mm x1 (Mid 2023),

Link to comment
Share on other sites
Link to post
Share on other sites
Posted

GP 100 = 980ti + 1000 cuda cores+ HBM2... and thats about it?

Intel Core i7 7800x @ 5.0 Ghz with 1.305 volts (really good chip), Mesh OC @ 3.3 Ghz, Fractal Design Celsius S36, Asrock X299 Killer SLI/ac, 16 GB Adata XPG Z1 OCed to  3600 Mhz , Aorus  RX 580 XTR 8G, Samsung 950 evo, Win 10 Home - loving it :D

Had a Ryzen before ... but  a bad bios flash killed it :(

MSI GT72S Dominator Pro G - i7 6820HK, 980m SLI, Gsync, 1080p, 16 GB RAM, 2x128 GB SSD + 1TB HDD, Win 10 home

 

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
On 06/04/2016 at 9:05 PM, Mr_Troll said:

GP 100 = 980ti + 1000 cuda cores+ HBM2... and thats about it?

And it's a 43% increase in FP32 performance over the Titan X, which is the most important metric.  And the P100 is also not Big Pascal.  It's missing 4 SM units, which is 256 extra CUDA cores.  Every new architecture, 1 CUDA core does more than the previous 1 CUDA core. 1 Maxwell CUDA core =/= 1 Pascal CUDA core.

QUOTE ME IN A REPLY SO I CAN SEE THE NOTIFICATION!

When there is no danger of failure there is no pleasure in success.

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
18 minutes ago, Samfisher said:

And it's a 43% increase in FP32 performance over the Titan X, which is the most important metric.  And the P100 is also not Big Pascal.  It's missing 4 SM units, which is 256 extra CUDA cores.  Every new architecture, 1 CUDA core does more than the previous 1 CUDA core. 1 Maxwell CUDA core =/= 1 Pascal CUDA core.

A CUDA core is just a stream processor. It's the clusters of them which change in quality and characteristics. The big thing for Nvidia in Maxwell was Delta Color Compression and far better ROPs and TMUs. The SPs may have slightly altered designs, but all they are, are floating point units.

Software Engineer for Suncorp (Australia), Computer Tech Enthusiast, Miami University Graduate, Nerd

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
3 minutes ago, patrickjp93 said:

A CUDA core is just a stream processor. It's the clusters of them which change in quality and characteristics. The big thing for Nvidia in Maxwell was Delta Color Compression and far better ROPs and TMUs. The SPs may have slightly altered designs, but all they are, are floating point units.

I know.  Never said it was.  Was comparing floating point performance between Maxwell and Pascal, and that the P100 isn't the full blown Pascal chip.

QUOTE ME IN A REPLY SO I CAN SEE THE NOTIFICATION!

When there is no danger of failure there is no pleasure in success.

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
42 minutes ago, Samfisher said:

And it's a 43% increase in FP32 performance over the Titan X, which is the most important metric.  And the P100 is also not Big Pascal.  It's missing 4 SM units, which is 256 extra CUDA cores.  Every new architecture, 1 CUDA core does more than the previous 1 CUDA core. 1 Maxwell CUDA core =/= 1 Pascal CUDA core.

P100 uses the GP100, the top tier Pascal GPU. I highly doubt they'll bin it for their HPC clients... they didn't do that for GM200 AFAIK

Spartan 1.0

Spoiler

CPU: Intel Core i7-4770K 3.5GHz Quad-Core Processor

CPU Cooler: Cooler Master Seidon 120XL 86.2 CFM Liquid CPU Cooler

Motherboard: Asus Maximus VI Extreme ATX LGA1150 Motherboard
Memory: Corsair Dominator 32GB (4 x 8GB) DDR3-1600 Memory
Storage: OCZ Vector Series 512GB 2.5" Solid State Drive
Storage: Seagate Desktop HDD 4TB 3.5" 7200RPM Internal Hard Drive

Video Card: EVGA GeForce GTX 980 4GB Classified ACX 2.0 Video Card
Case: Thermaltake Urban S41 ATX Mid Tower Case
Power Supply: Corsair 1200W 80+ Platinum Certified Fully-Modular ATX Power Supply
Optical Drive: LG BH16NS40 Blu-Ray/DVD/CD Writer
Optical Drive: LG BH10LS30 Blu-Ray/DVD/CD Writer
Operating System: Microsoft Windows 10 Pro 64-bit
Sound Card: Creative Labs ZXR 24-bit 192 KHz Sound Card
Monitor: 2x Asus VG278HE 27.0" 144Hz Monitor
Keyboard: Logitech G19s Wired Gaming Keyboard
Keyboard: Razer Orbweaver Elite Mechanical Gaming Keypad Wired Gaming Keyboard
Mouse: Logitech G700s Wireless Laser Mouse
Headphones: Creative Labs EVO ZxR 7.1 Channel  Headset
Speakers: Creative Labs GigaWorks T40 Series II 32W 2ch Speakers

Hades 1.0

Spoiler

Laptop: Dell Alienware 15 2015

CPU: i7-4720HQ CPU

Memory: 16GB DDR3 SODIMM RAM

Storage: 256GB M.2 SSD

Storage: 1TB 5400rpm 2.5" HDD

Screen: 15.6" FHD Display

Video Card: Nvidia GTX 970M with 3GB

Operating System: Windows 10 Pro

Project: Spartan 1.2 PLEASE SUPPORT ME NEW CHANNEL > Tech Inquisition

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
1 minute ago, GidonsClaw said:

P100 uses the GP100, the top tier Pascal GPU. I highly doubt they'll bin it for their HPC clients... they didn't do that for GM200 AFAIK

Like I mentioned in the other thread, Nvidia's devblog supports what I said :P

 

Quote

GP100 achieves its colossal throughput by providing six GPCs, up to 60 SMs, and eight 512-bit memory controllers (4096 bits total). Each SM has 64 CUDA cores and four texture units, for a total of 3840 CUDA cores and 240 texture units.

P100 is 56 SMs and therefore isn't Big Pascal. P100 has 3584 CUDA cores.  GP100 will have 3840.  That's the Titan class GPU we will be getting for desktops.

QUOTE ME IN A REPLY SO I CAN SEE THE NOTIFICATION!

When there is no danger of failure there is no pleasure in success.

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
1 minute ago, Samfisher said:

Like I mentioned in the other thread, Nvidia's devblog supports what I said :P

 

P100 is 56 SMs and therefore isn't Big Pascal. P100 has 3584 CUDA cores.  GP100 will have 3840.  That's the Titan class GPU we will be getting for desktops.

And like I replied... Ok... let's for arguments sake agree this is a binned GP100... possibly the next Ti?

Spartan 1.0

Spoiler

CPU: Intel Core i7-4770K 3.5GHz Quad-Core Processor

CPU Cooler: Cooler Master Seidon 120XL 86.2 CFM Liquid CPU Cooler

Motherboard: Asus Maximus VI Extreme ATX LGA1150 Motherboard
Memory: Corsair Dominator 32GB (4 x 8GB) DDR3-1600 Memory
Storage: OCZ Vector Series 512GB 2.5" Solid State Drive
Storage: Seagate Desktop HDD 4TB 3.5" 7200RPM Internal Hard Drive

Video Card: EVGA GeForce GTX 980 4GB Classified ACX 2.0 Video Card
Case: Thermaltake Urban S41 ATX Mid Tower Case
Power Supply: Corsair 1200W 80+ Platinum Certified Fully-Modular ATX Power Supply
Optical Drive: LG BH16NS40 Blu-Ray/DVD/CD Writer
Optical Drive: LG BH10LS30 Blu-Ray/DVD/CD Writer
Operating System: Microsoft Windows 10 Pro 64-bit
Sound Card: Creative Labs ZXR 24-bit 192 KHz Sound Card
Monitor: 2x Asus VG278HE 27.0" 144Hz Monitor
Keyboard: Logitech G19s Wired Gaming Keyboard
Keyboard: Razer Orbweaver Elite Mechanical Gaming Keypad Wired Gaming Keyboard
Mouse: Logitech G700s Wireless Laser Mouse
Headphones: Creative Labs EVO ZxR 7.1 Channel  Headset
Speakers: Creative Labs GigaWorks T40 Series II 32W 2ch Speakers

Hades 1.0

Spoiler

Laptop: Dell Alienware 15 2015

CPU: i7-4720HQ CPU

Memory: 16GB DDR3 SODIMM RAM

Storage: 256GB M.2 SSD

Storage: 1TB 5400rpm 2.5" HDD

Screen: 15.6" FHD Display

Video Card: Nvidia GTX 970M with 3GB

Operating System: Windows 10 Pro

Project: Spartan 1.2 PLEASE SUPPORT ME NEW CHANNEL > Tech Inquisition

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
Just now, Samfisher said:

Like I mentioned in the other thread, Nvidia's devblog supports what I said :P

 

P100 is 56 SMs and therefore isn't Big Pascal. P100 has 3584 CUDA cores.  GP100 will have 3840.  That's the Titan class GPU we will be getting for desktops.

Indeed. For NVidia to use cut down chips for their high end business market should tell you something about their yields at the moment. A high end consumer model is far away for sure.

Watching Intel have competition is like watching a headless chicken trying to get out of a mine field

CPU: Intel I7 4790K@4.6 with NZXT X31 AIO; MOTHERBOARD: ASUS Z97 Maximus VII Ranger; RAM: 8 GB Kingston HyperX 1600 DDR3; GFX: ASUS R9 290 4GB; CASE: Lian Li v700wx; STORAGE: Corsair Force 3 120GB SSD; Samsung 850 500GB SSD; Various old Seagates; PSU: Corsair RM650; MONITOR: 2x 20" Dell IPS; KEYBOARD/MOUSE: Logitech K810/ MX Master; OS: Windows 10 Pro

Link to comment
Share on other sites
Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share
Go to topic listing Tech News

×